JPS62239285A - Central line extracting device for slit image - Google Patents

Abstract

PURPOSE:To extract the central line of a slit image with high accuracy and at high speed even when a noise is included in the slit image by obtaining the gravity center of the brightness distribution of the slit image and making this into the central point of the slit image. CONSTITUTION:A projector 9 irradiates a slit light 6 to an observing object 7, generates an irradiating pattern 8 of the slit light at the surface of the observing object 7, and a television camera 10 photographs the irradiating pattern 8 and generates a slit image. Into a data storing memory 11, the set of a position data J to specify the width direction position of the slit image and two component data I.J to make brightness data I to specify the brightness in respective positions into a component are stored. An accumulating arithmetic circuit 13 accumulates and calculates the data accessed by a data address generator 12, and a computer calculates the gravity center position of the brightness distribution of the slit image as a point to constitute the central line of the slit image.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention irradiates an object to be observed with slit light to generate a slit light irradiation pattern on its surface, and images the irradiation pattern to create a slit image. The present invention relates to an object recognition technology that recognizes the position and shape of the observation target by analyzing the image after the image is obtained, and in particular, the present invention relates to an object recognition technology that recognizes the position and shape of the observed object by analyzing the image. It is connected to a centerline extraction device for slit images, which is used for highly accurate extraction.

<Conventional technology> In this type of conventional center line extraction device, the 8th
A slit image 1 as shown in the figure is captured, and the brightness distribution 2 in the width direction of this slit image 1 (see Figure 9) is observed directly.
3, a point 5 (hereinafter referred to as "center point") constituting the center line 4 of the slit image 1 is extracted.

In order to extract this center point 5, the position where the highest brightness is observed in the brightness distribution 2 (in Fig. 9, the point P
) and using this as the center point, or the range d of positions where brightness above a certain threshold value TH is observed.
The method used is to calculate the average value and use this as the center point.

<Problems to be Solved by the Invention> However, in the case of using the former method, if the slit image 1 contains a lot of noise (for example, spectrum noise), the maximum brightness may not be achieved due to the noise. Therefore, it is difficult to always extract a proper center point.

In addition, when using the latter method, since different objects to be observed have different reflectances, etc., it is necessary to change the threshold value Tt() for each object to be observed. There are problems such as it is not easy to make decisions.

In view of the above-mentioned circumstances, this invention makes it possible to extract the center line of a slit image with high accuracy and high speed by employing a method of finding the center of gravity of the brightness distribution of the slit image and using this as the center point of the slit image. Moreover, it is an object of the present invention to provide a center line extraction device for a slit image that does not require any threshold setting or the like.

<Means for Solving the Problems> The structure of the present invention for achieving the above object will be explained using FIGS. 1 to 3 corresponding to one embodiment. The line extraction device uses a light projecting means (
(corresponds to the light projector 9 in the figure), imaging means (corresponds to the television camera 10 in the figure), data storage means (corresponds to the data storage memory 11 in the figure), data access means (corresponds to the data storage memory 11 in the figure),・Corresponding to the address generator 12), cumulative calculation means (corresponding to the cumulative calculation circuit 13 in the same figure), and 1 center of gravity position calculation means (corresponding to the computer 1 in the same figure)
4).

The light projecting device 9 is for irradiating the observation object 7 with the slit light 6 to generate an irradiation pattern 8 of the slit light on the surface of the observation object 7, and also for projecting the slit light 6 onto the observation object 7.
0 generates a slit image by imaging the irradiation pattern 8 of the slit light.

The data storage memory 11 stores two-component data (the components of which are a north position data J that defines the width direction position of the slit image and brightness data I that defines the brightness at each position).
1, J) are stored, and the data address generator 12 is for sequentially accessing a plurality of data stored in the data storage memory 11.

The cumulative calculation circuit 13 performs cumulative calculation on the data accessed by the data address generator 12.
The combiator 14 is used to calculate the center of gravity of the brightness distribution of the slit image as a point forming the center line of the slit image using the calculation result of the cumulative calculation circuit 13.

Effect> When the observation object 7 is irradiated with the slit light 6 by the light projection device 9, an irradiation pattern 8 of the slit light is generated on the surface of the observation object 70 according to the surface shape of the object. When the irradiation pattern 8 of this slit light is imaged by the television camera 10, the above-mentioned! ! A slit image having brightness corresponding to the brightness of the q-ray pattern 8 is generated.

The brightness distribution in the width direction of this slit (elephant) is obtained along the observation straight line 3 (see FIG. 8), and the position data J defining the width direction position of the slit image 1 and each A plurality of two-component data J) whose component is brightness data I that defines brightness at a position are stored.

The data stored in this data storage memory 11 is sequentially accessed by a data address generator 12, and the accessed data is input to an accumulation calculation circuit 13 where predetermined calculations such as product-sum calculations and accumulation calculations are executed. Ru. The calculation result by the cumulative calculation circuit 13 is sent to the computer 14.
This calculation result is used to calculate the center of gravity of the brightness distribution of the slit image, and the calculated value is extracted as the center point of the slit image.

A similar center point extraction process is repeatedly executed over the entire length of the slit image, thereby extracting the center line of the slit image.

<Embodiment> FIG. 3 shows the overall schematic configuration of an object recognition device incorporating the center line extraction device of the present invention.

The illustrated device example includes a light projection device 9 that irradiates an observation object 7 with slit light 6 to generate a slit light irradiation pattern 8 on the surface thereof, and a light projection device 9 that images this irradiation pattern 8 and displays a slit light beam on the imaging surface. It includes a television camera 10 made of a two-dimensional CCD that generates a slit image 1 as shown in FIG. A device is used in combination with a cylindrical lens.

A control processing device 15 is connected to the projector 9 and the television camera 10, and this control processing device 15 takes in the slit image 1 taken by the television camera 10, extracts the center line of the slit image 1, and extracts the object. Perform image analysis for recognition.

1 and 2 show examples of circuit configurations for extracting the center line of the slit image 1.
is captured and stored in the image memory 16 in FIG.

FIG. 5 shows an enlarged view of the slit image 1 in the image memory 16. In the figure, the X-axis and the y-axis indicate pixel positions in the image memory 16, and when the observation straight line 3 for obtaining the brightness distribution in the width direction of this slit image l is set as shown in the figure. , J a "" J N-1 is obtained as position data for defining the position of the slit image 1 in the width direction.

FIG. 6 shows the brightness distribution of the slit image 1 along the observation straight line 3, and in the figure, JK is the brightness distribution of the slit image 1.
I(JX)
is position data J.

Brightness data that defines the brightness of the slit image 1 at the bitter position is shown.

In this invention, the position of the center of gravity of the brightness distribution of the slit image 1 (the position indicated by position data J in FIG. 5.6) is calculated by the following (2), and the center point of the slit image 1 is That is.

An example of the configuration of an arithmetic circuit T that is required to execute at high speed with Jc=□...■ is shown, and FIG. 4 shows a timing chart of the circuit shown in FIG.

The illustrated example circuit includes a data storage memory 11, a data address generator 12, an accumulation operation circuit B13, and a timing generator 17.

The data storage memory 11 is for storing two-component data whose components are position data J that defines the position in the width direction of the slit image I and brightness data I that defines the brightness at each position. In the figure, the address area indicated by 0 to N-1 contains N pieces of two-component data (1,, J,).
・” (IN-,, JN-6) is stored.

Data address generator No. 12 is used to output data addresses and sequentially access each data in the data storage memory 11.

The cumulative calculation circuit 13 consists of a product-sum calculation unit L8 and an accumulator 19, and each time the data accessed by the data address generator 12 is input, the product-sum calculation unit 18 performs a product-sum calculation, respectively. Run in parallel.

The timing generator 17 receives various control signals INCP, ACC,
This is for generating END and MAP. Among these, INCP is an address increment signal requesting address update, and upon input of this signal, the data address generator 12 increments by 1. ACC is a logic processing signal that controls the start of the product-sum processing of the accumulation calculation circuit 13. EN
D is a data manual signal, which instructs input of data from the data storage memory 11. MAP is a product-sum signal,
Controls execution of product-sum processing by the cumulative calculation circuit 13.

In the figure, MA indicates the start address of the data string, MD indicates 1. Each component data of J is shown below. Further, ST is a start signal, which instructs the start of hardware processing in this circuit. CADR is the final address access signal,
Notify that the data start address has become the end address. END is an end signal that instructs the end of hardware processing in this circuit.

The circuit shown in FIG. 1 includes the above-mentioned data storage memory 11, data address generator 12, and accumulation calculation circuit
, a timing generator 17, a register 20 in which a data string end address is set, and a comparator that compares the data string start address updated by the data address generator 12 with the data string end address. - Contains evening 21.

FIG. 2 shows the overall configuration of a center line extraction device including the arithmetic circuit T.

In the figure, the combiator 14 accesses the image memory [6, and brightness data 1 (
JK) and stores it in the data storage memory 11 of the arithmetic circuit T.
Store in. Further, when the data storage process is completed, the computer 14 sends the start signal ST to the arithmetic circuit T to start the cumulative arithmetic process, and when the computer 14 receives an end signal END from the arithmetic circuit T, the computer 14 uses the arithmetic result of the arithmetic circuit T. The center point of the slit image 1 is calculated by executing the calculation in (2) above.

Next, the operation of the circuit shown in FIG. 1 will be explained with reference to the timing chart shown in FIG.

First, when the data address generator 12 receives a start signal ST from the computer 14, it accesses the data storage memory 11 using a preset data string start address MA. During states O to 4 (one clock cycle), the entire circuit waits for the data MD constituting the data string to be accessed.

When state 5 is reached, the input section of the cumulative calculation circuit 13 has:
Data input signal END from timing generator 17
is input, and together with it, 1, J, which constitutes a data string
Each component data MD is input manually from the data storage memory IL.

At the same time, the address increment signal INCP rises, which causes the data address generator 12
updates the start address MA of the data string by adding 1, and accesses the data storage memory 11 with the new start address MA. While this data storage memory 11 is being accessed, the timing generator 17 is in state 7.
A sum-of-products signal MAP is generated, and necessary component data is loaded into the sum-of-products calculator 18 and accumulator 19 of the cumulative calculation circuit section 13. That is, the product-sum calculator 18 receives component data Io and J.

However, the accumulator 19 also contains component data [. are loaded respectively. At this point, the product-sum processing signal ACC is not active, so the accumulation calculation circuit 13 does not perform any calculation. When the product-sum processing signal ACC becomes active in state 8, the accumulation calculation circuit 13 The summation unit 18 and the accumulator 19 start a predetermined accumulation calculation process.

Next, in state 9, the input section of the cumulative calculation circuit 13 has the following information:
When the data input signal IND is input, the next data MD constituting the data string is input from the data storage memory 11, and at the same time, the address increment signal INCP rises and the data string start address MA is updated by adding 1. .

Then, in state 11, the sum of products signal MA is sent to the cumulative calculation circuit 13.
P is input, and necessary component data is loaded into the product-sum calculator 18 and accumulator 19, and the product-sum calculator 18 and accumulator t9 perform predetermined cumulative calculation processing (product-sum calculation processing). The arithmetic unit 18 starts the operations (2). Jo+11 J+, and the accumulator 19 starts Io+It.

Thus, the comparator 21 determines the data string start address M
Until the match between A and the end address is determined and the end address access signal CADR is output to the timing generator 17, the process repeatedly returns to state 8 and performs the same cumulative arithmetic operation. and the state [
When it is detected in state 1 that the final address access signal CADR is active, the next state 12 is entered. In the following state 13, the next data constituting the data string is input to the cumulative calculation circuit section 13, and further, in state 15, a predetermined cumulative calculation is executed. The cumulative calculation is completed, and at this point, the timing generator 17 sends an end signal END to the computer 14. When this end signal END is output, the start signal ST becomes active and the timing generator 17 returns to state 0.

When the cumulative calculation of all the data is completed in the calculation circuit T shown in FIG. The position of the center of gravity of the brightness distribution of the slit image 1 is calculated as the center point of the slit image 1.

A similar center point extraction process is repeatedly executed over the entire length of the slit image 1, thereby making it possible to extract the center line of the slit image 1.

<Effects of the Invention> As described above, this invention employs a method of finding the center of gravity of the brightness distribution of the slit image and using this as the center point of the slit image, so even if the slit image contains noise, The center line of the slit image can be extracted with high precision and at high speed, and there is no need to set a threshold value as in the conventional example, which is a remarkable effect that achieves the purpose of the invention.

[Brief explanation of drawings]

1 and 2 are block diagrams showing an example of the circuit configuration of a device for detecting the center line of an elephant according to an embodiment of the present invention, and FIG. Fig. 4 is an explanatory diagram showing the schematic configuration, Fig. 4 is a timing chart of the circuit in Fig. 1, Fig. 5 is an explanatory diagram showing a state in which an observation straight line is set on the slit image, and Fig. 6 is an explanatory diagram showing the state in which the observation straight line is set on the slit image. An explanatory diagram showing the brightness distribution, Fig. 7 is an explanatory diagram showing the format of data stored in the data storage memory, Fig. 8 is an explanatory diagram showing the center line of the slit image, and Fig. 9 is an explanatory diagram showing the conventional slit image. It is an explanatory diagram showing a center line extraction method. 1... Slit image 3... Observation straight line 4...
... Center line 5 ... Center point 6 ... Slit light 8 ... Irradiation pattern 9 ... Light projector 10 ... Television camera 11 ... Data storage memory 12 ... ...Data address generator 13...
...Accumulative calculation circuit 14...Computer patent
Applicant: Tateishi Electric Co., Ltd. -1 - φ Hexagram / +zl - Su J color Iku 1 no 1 conu b difference 48 F mouth, 21z7F2) '13 - 4 draw color f row - B multi-account - θ entry Male block + 21 bones 3 figures

Claims (5)

[Claims] (1) Light projecting means for irradiating an observation object with slit light to generate an irradiation pattern of the slit light on the surface of the observation object, and generating a slit image by imaging the irradiation pattern of the slit light. an imaging means; a data storage means for storing two-component data whose components are position data that defines the position in the width direction of the slit image and brightness data that defines the brightness at each position; A data access means for sequentially accessing stored data; an accumulation calculation means for cumulatively calculating the data accessed by the data access means; and a brightness distribution of the slit image using the calculation results of the accumulation calculation means. A centerline extraction device for a slit image, comprising a centerline calculation means for calculating the centerline of the slit image as constituent points of the centerline of the slit image. (2) The slit image centerline extraction device according to claim 1, wherein the imaging means is a television camera comprising a two-dimensional CCD. (3) The data access means includes a data address,
The slit image centerline extraction device according to claim 1, which is a generator. (4) The slit image centerline extraction device according to claim 1, wherein the cumulative calculation means comprises a product-sum calculation unit that performs a product-sum calculation and an accumulator that performs an accumulation calculation. (5) The slit image center line extraction device according to claim 1, wherein the gravity center position calculation means is a computer.